Machine vision 3D line scan image acquisition and processing

1. A processing device for processing a binary measurement signal, the processing device comprising: a first buffer to receive the binary measurement signal read out from a pixel array with N 1 rows and N 2 columns, wherein the binary measurement signal corresponds to an optical image formed at the pixel array, wherein the binary measurement signal has been read out from the pixel array as a matrix with M rows and N 2 columns, where M is smaller than N 1 , by sampling analog column output signals representative of the optical image and formed on columns of the pixel array based on a sampling matrix factored as a product of (i) a sparse random matrix with M rows and N 1 columns and (ii) a spatial filtering function corresponding to a partial first derivative with respect to the rows of the pixel array; a second buffer to store the sparse random matrix; first digital processing circuitry arranged and configured to perform a sparse matrix multiplication of a transposed instance of the sparse random matrix with the binary measurement signal; second digital processing circuitry arranged and configured to convolve a result of the sparse matrix multiplication with a smoothing kernel, that is related to the spatial filtering function corresponding to the partial first derivative with respect to the rows of the pixel array, to obtain a N 1 ×N 2 digital image corresponding to the optical image effectively filtered with the spatial filtering function corresponding to the partial first derivative with respect to the rows of the pixel array; and a third buffer to store the digital image for transmission, display or further processing.

2. The processing device of claim 1 , comprising: a third digital processing circuitry arranged and configured to determine peaks in columns of the obtained digital image; and calculate offset parameters associated with points of interest in the optical image based in the determined peaks.

3. The processing device of claim 1 , wherein the first digital processing circuitry comprises: an array of multipliers arranged and configured to receive a row of the binary measurement signal from the first buffer, and multiply the row with a corresponding coefficient of a row of the transpose instance of the sparse random matrix; an array of accumulation registers; and an array of adders arranged and configured to add values output by the array of multipliers to corresponding values accumulated in the array of accumulation registers from previous multiplications of other rows of the binary measurement signal with respective other coefficients of the row of the transpose, wherein a row of the digital image corresponding to the row of the transpose is formed as values accumulated in the array of accumulation registers after multiplications of all the rows of the binary measurement signal by all coefficients of the row of the transpose.

4. A vision system comprising: a digital camera comprising a pixel array including a plurality of pixel elements partitioned into N 1 rows and N 2 columns, each of the pixel elements comprising a light sensor; wherein the N 2 pixel elements in each of the N 1 rows have common control couplings for receiving control signals for pixel elements of that row, the common control couplings of each of the N 2 pixel elements include a first control coupling for receiving a control signal indicative of a first of two bits of information and a second control coupling for receiving a control signal indicative of a second of the two bits of information, wherein the control signals are coefficients of a sampling matrix factored as a product of (i) a sparse random matrix with M rows and N 1 columns, where M is smaller than N 1 , and (ii) a spatial filtering function corresponding to a partial first derivative with respect to the rows of the pixel array; wherein the N 1 pixel elements in each of the N 2 columns have common output couplings for providing each of a first analog column output signal on a first output coupling and a second analog column output signal on a second output coupling for that column, the first analog column output signal being dependent on light energy accumulated by two or more of the N 1 pixel elements of that column that are selectively coupled with respective first control couplings that have received respective control signals indicative of the first of the two bits of information, and the second analog column output signal being dependent on light energy accumulated by a different two or more of the N 1 pixel elements of that column that are selectively coupled with respective second control couplings that have received respective control signals indicative of the second of the two bits of information, the light energy accumulated on the N 1 ×N 2 pixel elements of the pixel array being from an optical image formed at the pixel array, wherein the two bits of information represent which, if any, of the first and second output couplings is to be selected; and an array of comparators coupled in one-to-one correspondence with the columns of the pixel array, each of the N 2 comparators to output a one-bit value that indicates whether or not a magnitude of the first analog column output signal is larger than a magnitude of the second analog column output signal from its corresponding column of the pixel array, the N 2 one-bit values output by the array of comparators being a row of a binary measurement signal, wherein the binary measurement signal corresponds to the optical image formed at the pixel array, the binary measurement signal being a matrix with M rows and N 2 columns; and a digital computer coupled with the digital camera via a digital communication channel, the digital computer comprising a processor and a memory device encoding instructions that, when executed by the processor, cause the digital computer to perform operations comprising receiving, on a row-by-row basis, the binary measurement signal from the digital camera; performing a sparse matrix multiplication of a transposed instance of the sparse random matrix with the binary measurement signal; convolving a result of the sparse matrix multiplication with a smoothing kernel, that relates to the spatial filtering function corresponding to the partial first derivative with respect to the rows of the pixel array, to obtain a N 1 ×N 2 digital image corresponding to the optical image effectively filtered with the spatial filtering function corresponding to the partial first derivative with respect to the rows of the pixel array; and outputting the digital image for transmission, display or further processing.

5. The vision system of claim 4 , wherein the digital camera comprises a shift register coupled with the control couplings of the respective rows of the pixel array, wherein the shift register is configured to receive control signals that form a row of the sampling matrix, and provide the received control signals to the control couplings of the respective rows of the pixel array.

6. The vision system of claim 4 , wherein the operations further comprise determining peaks in columns of the obtained digital image; and calculating offset parameters associated with points of interest in the optical image based in the determined peaks.

7. The vision system of claim 4 , wherein the digital camera comprises a storage element associated with the control couplings of the rows, wherein the storage element comprises an array of storage cells, each of the storage cells including storage for the two bits of information representing the control signals.

8. The vision system of claim 7 , wherein the digital camera comprises a shift register having an input coupled with the storage element and outputs coupled in one-to-one correspondence with the control couplings of the respective rows of the pixel array, wherein the shift register is configured to receive a number of control signals stored in the storage element that is equal to the number of rows of the pixel array, and provide the received control signals to the control couplings of the respective rows of the pixel array.

9. The vision system of claim 8 , wherein the digital camera comprises an input buffer having an input coupled to the storage element and an output coupled to the shift register, wherein the input buffer is configured to transfer the control signals from the storage element to the shift register in subsets of the number of control signals.

10. The vision system of claim 9 , wherein the shift register is configured to retain previously received control signals while new control signals are being transferred by the input buffer from the storage element to the shift register, thereby maintaining a state of the control couplings while new control signals are being transferred to the shift register.

11. The vision system of claim 4 , wherein the digital camera comprises: a first storage element disposed adjacent one side of the pixel array for storing information representing the control signals; and a second storage element disposed adjacent another side of the pixel array for storing a copy of the information representing the control signals.

12. The vision system of claim 4 , wherein the light sensor of at least some of the pixel elements comprises a pinned photodiode.

13. The vision system of claim 4 , wherein each of the pixel elements comprises a transconductor, and a floating diffusion node coupled with the light sensor and the transconductor, the floating diffusion node configured to store charge accumulated by the light sensor, such that a voltage of the floating diffusion node causes the transconductor to provide a pixel current proportional to the voltage.

14. The vision system of claim 13 , wherein each of the pixel elements comprises a first transistor coupled with the transconductor, the control coupling of the corresponding row, and the first output coupling of the corresponding column, and a second transistor coupled with the transconductor, the control coupling of the corresponding row, and the second output coupling of the corresponding column, the first transistor and the second transistor are arranged and configured to switch the pixel current provided by the transconductor to the first output coupling if the control coupling has received a control signal indicative of the first of the two bits of information, thereby the pixel current to contribute to the first analog column output signal on the first output coupling, and the second output coupling if the control coupling has received a control signal indicative of the second of the two bits of information, thereby the pixel current to contribute to the second analog column output signal on the second output coupling.

15. The vision system of claim 4 , wherein the digital camera comprises: an output data multiplexer coupled through 1-bit digital output signal lines with respective outputs of the comparators, wherein the output data multiplexer is configured to output the row of the binary measurement signal.